The proposed project will examine trajectories, consequences, and multiple levels of influences on alcohol use among urban poor adolescents, explicitly comparing patterns of effects across ethnic and gender subgroups. The long-term outcome of the proposed project is improved preventive interventions with larger beneficial effects efficiently achieved across the diverse population of U.S. youth. Current integrative theoretical models propose that behavior is influenced by a wide array of macro/distal to micro/proximal personal, social and environmental factors, yet few studies to date have included variables at multiple levels and from multiple domains, as proposed here, permitting a comprehensive study of the etiology of onset and trajectories of alcohol use among multi-ethnic groups of adolescents. Study aims include: (1) improve understanding of multilevel factors on trajectories of alcohol use by ethnic and gender subgroups, (2) compare effects of age of onset of alcohol use and age of initial intoxication on alcohol use at age 18 across ethnic and gender subgroups, (3) identify similarities or differences in consequences of alcohol use across ethnic and gender subgroups, and (4) assess the long-term effectiveness of the Project Northland Chicago (PNC) intervention and its differential effectiveness across subgroups. Methods to achieve the study aims include: (1) cost- effective utilization of existing longitudinal student, parent, school, and neighborhood-level data, including repeated annual surveys from a large majority-minority cohort ages 11 to 14 that have already been collected by the investigators in a previous major alcohol prevention trial;(2) implementation of a mail and web-based follow-up survey of the study cohort when they are age 17-18;and (3) advanced, multilevel, longitudinal statistical analyses. The cohort includes adolescents who completed at least one previous PNC survey when they were in the beginning and end of 6th grade, end of 7th grade, and end of 8th grade (n = 5,812). The study cohort attend public schools in the city of Chicago, are mostly low SES (72 percent), and primarily Black (43 percent), Hispanic (28 percent), and White (13 percent). PUBLIC HEALTH RELEVANCE: Results will significantly improve understanding of alcohol use trajectories among Black and Hispanic youth, and will guide development of theory applicable to ethnic and gender subgroups. Most importantly, results will directly guide the development of further refined interventions of increased efficacy and effectiveness.